Gamma band (30-90 Hz) oscillations are hypothesized to play an important role in normal cognition, including memory encoding and retrieval, attention and perception. Gamma synchrony is abnormally regulated in many disorders, such as epilepsy, schizophrenia and dementias such as Alzheimer's disease. Distinct mechanisms likely underlie gamma oscillations in different brain areas, and mechanisms may also vary within the same brain area under different conditions. Models of these diverse mechanisms generally assume that interneurons function as integrators that can fire at arbitrarily low rates (type 1 excitability). In contrast, resonator neurons have an abrupt threshold at a nonzero minimum firing frequency (type 2 excitability). We have previously shown that the fast spiking (FS), parvalbumin-positive (PV+) basket cell interneurons in the medial entorhinal cortex (MEC) are type 2, and exhibit strong resonance and post-inhibitory rebound (PIR). Moreover, our theo- retical work shows these features enhance the ability to synchronize in heterogeneous, sparsely-connected noisy networks. Aim 1 will focus on the biophysical basis for PIR and type 2 excitability in FS cells in mouse MEC and hippocampal area CA3 in vitro. Aim 2 will use models of CA3 and MEC FS cells from Aim 1 embed- ded in excitatory/inhibitory networks to develop new theory to identify and optimally manipulate the various mechanisms underlying gamma synchrony. We will analyze different slices of the parameter space to find or- ganizing principles for distinct gamma mechanisms and how to distinguish between them. We will develop the- oretical methods to account for the effect of jitter in spike times. This theory may lead to better design of poten- tial therapies for cognitive deficits. Aim 3 will test the theoretical predictions of optimally gated transitions into theta-nested gamma in the MEC in vitro using optogenetic control of extrinsic inputs. We will test the hypothe- ses that both excitatory and inhibitory theta-locked signals can evoke nested gamma oscillations during opto- genetically-induced theta in the MEC by aligning the phases of the FS interneurons. A consistent reset of the theta phase of gamma oscillations is required in many coding schemes; we expect that multiple reset mecha- nisms may be operative in the MEC. Our central hypothesis is that the excitability type of inhibitory interneu- rons controls the type and robustness of oscillations exhibited in excitatory/inhibitory networks.